A frequent goal of signal processing is to improve the quality, or the fidelity, of a captured signal to the information it represents by reducing noise in the signal. For example, recorded audio signals are often processed to remove noise and other undesirable signal components to create an audio signal much more similar to the original sound that was recorded. However, conventional methods used to reduce noise are sometimes ineffective or slow. In some methods, the level of noise reduction is unsatisfactory. In other methods, the reduction of noise is destructive and removes a significant amount of desired information from the signal. In addition, many of these methods require an excessive amount of processing time to perform, tying up processing resources.
These conventional noise filtering methods are often utilized in digital imaging applications, such as photographic film digitization, to reduce noise caused by film grain or noise introduced by the image capturing equipment. Many conventional noise filtering methods for images utilize blurring of the base image to reduce noise. However, the use of blurring on the base image often causes a degradation of detail, as the edges are blurred. To prevent a significant loss of detail, conventional image noise reduction methods reduce the level of noise reduction, thereby diminishing the effectiveness of the blurring process. The conventional blurring methods can also require relatively extensive processing as the base image is processed at the base resolution.
Given the drawbacks in current noise reduction methods, it is clear that conventional methods of improving signal quality are less than perfect.